why?

Do you love lifting weights and HATE doing plate math? I certainly do. So, instead of trying to add and subtract and compute percentages, let's pre-compute our warmup sets instead!

example html

335

1 warmup set

set	total	side	add
e	45
A	185	70	45, 25
W	335	145	35, 15, 10, 5, 5, 5

2 warmup sets

set	total	side	add
e	45
A	135	45	45
B	235	95	35, 15
W	335	145	25, 10, 5, 5, 5

3 warmup sets

set	total	side	add
e	45
A	115	35	35
B	185	70	25, 10
C	275	115	45
W	335	145	15, 5, 5, 5

how to

pip install tabulate, install pandoc for html rendering: https://pandoc.org/installing.html

if needed, change in main.py:

  # how many on one side of the bar
  my_weights: Dict[Numeric, int] = {
    5: 3,
    10: 1,
    15: 1,
    25: 1,
    35: 1,
    45: 1
  }

run main.py then o.cmd

algorithm

the general idea is to evenly divide the jumps between 45 and the target weight and do a full search to see what combinations of adds and removes touch those intermediate steps. also, do +/- 5 to the steps (15 -> 10, 15, 20) because that's close enough and can sometimes result in a better path.

"better" (or "ranking") is done by:

1. the fewest # of moves
2. the lowest total weight touched (adds + removes)
3. the lowest amount of weight removed (slightly redundant)
4. the lowest "centeredness" to prefer heavier weights being closer to the center of the bar

(avg_gap, max_gap, min_gap, and mse are informational.) to some extent, this is just a personal preference. my goal is to avoid thinking about it, so I just picked this as the "canonical" solution.

example intermediate table for 155:

#	t	-	center	final	sequence	steps	avg_gap	max_gap	min_gap	mse
4	55	0	380.78	(25, 15, 10, 5)	[(25, 0), (15, 0), (10, 0), (5, 1)]	(25, 40, 55)	18.33	25	15	16.67
4	55	0	407.25	(25, 10, 15, 5)	[(25, 0), (10, 0), (15, 0), (5, 1)]	(25, 35, 55)	18.33	25	10	8.33
4	55	0	447.68	(15, 25, 10, 5)	[(15, 0), (25, 0), (10, 0), (5, 1)]	(15, 40, 55)	18.33	25	15	16.67
4	55	0	448.14	(25, 5, 15, 10)	[(25, 0), (5, 1), (15, 0), (10, 0)]	(25, 30, 55)	18.33	25	5	16.67
4	55	0	654.88	(15, 10, 5, 25)	[(15, 0), (10, 0), (5, 1), (25, 0)]	(15, 30, 55)	18.33	25	15	16.67
4	55	0	675.32	(15, 5, 10, 25)	[(15, 0), (5, 1), (10, 0), (25, 0)]	(20, 30, 55)	18.33	25	10	8.33
5	85	-15	599.59	(5, 35, 15)	[(5, 1), (15, 0), (-15, 0), (35, 0), (15, 0)]	(20, 40, 55)	18.33	20	15	8.33
6	65	-5	500.63	(10, 25, 15, 5)	[(10, 0), (5, 1), (-5, 1), (25, 0), (15, 0), (5, 1)]	(15, 35, 55)	18.33	20	15	8.33
6	65	-5	675.32	(15, 5, 10, 25)	[(15, 0), (5, 1), (10, 0), (5, 0), (-5, 0), (25, 0)]	(20, 35, 55)	18.33	20	15	0

hacks

bar weight is hardcoded to 45 and modifying to use kilograms is left as an exercise for the reader.

130 is modeled as the same solutions as 125 + adding a 2.5 at the end of the path: this reduces the search space and speeds things up considerably. (see compute_wrapper in main.py.) modifying to support 131, 132, 132.5, etc. is also left as an exercise for the reader.

it'd be really cool to color-code the output in renderer.py...

files

• main.py - main program, generates output.pkl (takes ~3 minutes)
  • tests.py - some unit tests
• renderer.py - input is output.pkl, displays markdown to the console
• o.cmd - runs renderer.py and then pandoc to generate html
• extra.html - extra CSS used by pandoc to align the HTML table more nicely and make printing easier
• output.html - generated output
• output.pdf - chrome print output (1 weight per page)
• output4.pdf - chrome print output (4 weights per page)

example renderer.py output:

# 335

## 1 warmup set

| set   |   total |   side | add                 |
|-------|---------|--------|---------------------|
| e     |      45 |        |                     |
| A     |     185 |     70 | 45, 25              |
| W     |     335 |    145 | 35, 15, 10, 5, 5, 5 |

## 2 warmup sets

| set   |   total |   side | add             |
|-------|---------|--------|-----------------|
| e     |      45 |        |                 |
| A     |     135 |     45 | 45              |
| B     |     235 |     95 | 35, 15          |
| W     |     335 |    145 | 25, 10, 5, 5, 5 |

## 3 warmup sets

| set   |   total |   side | add         |
|-------|---------|--------|-------------|
| e     |      45 |        |             |
| A     |     115 |     35 | 35          |
| B     |     185 |     70 | 25, 10      |
| C     |     275 |    115 | 45          |
| W     |     335 |    145 | 15, 5, 5, 5 |